How to get more tips as a rideshare driver

Rarely. The largest study of rideshare tipping ever run — a field experiment across the whole US Uber platform, published through the NBER — found that only about 16% of trips receive a tip, and that the majority of riders never tip at all in a given period. A small minority tip on nearly every trip and carry most of the total.

That number reframes the whole question. If 84% of your trips end with nothing, the money is not in persuading a $3 tipper to give $5. It is in converting some part of the silent majority into tipping at all. Advice aimed at increasing the size of a tip is aimed at the smaller half of the problem.

Honest caveat first: nearly all of the good tipping research is restaurant research. It is decades old, the samples are often small, and none of it was run in a car. What follows is the part that has been tested more than once and has a plausible reason to transfer.

Smiling. In a 1978 study of a cocktail lounge, a broad smile roughly doubled the amount received compared with a minimal one. It is one server and a small sample — but it is the single most-replicated direction in the literature, and it costs nothing.

A small personal mark. Rind and Bordia found that a server drawing a smiley face on the back of the check raised tips substantially. The mechanism people usually propose is that it converts a transaction into a brief personal exchange.

Being findable at the right moment. Gratitude has a half-life. A rider who felt looked after when you lifted their case into the boot is not the same person who opens the app the next morning.

They cut both ways, and this is the most misunderstood finding in tipping. Haggag and Paci studied default tip suggestions on New York taxi payment screens and found that raising the defaults increased the average tip among people who tipped and simultaneously increased the share of riders who tipped nothing.

A number that feels like the expected amount is an anchor. A number that feels like a demand is a reason to opt out entirely — and once someone opts out, they usually opt out completely rather than typing a smaller figure.

The practical read: on a screen you control, the leftmost or lowest suggestion is the one doing most of the work, because that is the one most people take. Setting it high does not raise the middle. It raises the exits.

There is no study. Not a weak study — none. Every list telling you that bottled water raises tips is repeating other lists.

That does not make it wrong. A clean car and a comfortable temperature plausibly feed the same warmth channel that smiling does, and they are cheap. What it means is that you should not spend real money on amenities expecting a measurable return, because nobody has measured one. If you are handing out water at $0.30 a bottle across 20 trips a day, that is $6 a day against an effect size of exactly unknown.

The same goes for the advice to ask for tips outright. Asking is known to work in charitable giving, but a rideshare rider is a captive audience in a stranger's car, and a verbal ask is the one version of this that can obviously backfire on your rating.

Because it is the one thing in this list that has not already been optimised by the platform. You cannot change how Uber's tip screen looks. You can change whether a grateful rider has any way to act while they are still grateful.

This is the reasoning behind in-car tipping prompts generally — signs, QR codes, NFC tags. A rider who is reaching for the door handle and feels they had a good ride currently has to remember, find the app, and act on that feeling several hours later. Most don't.

It is also the thing this product exists to do, so treat that paragraph with the scepticism it deserves — we have a commercial interest in you believing it. The part that is not our opinion is the 16% figure above, and the fact that gratitude decays.